You are on page 1of 5

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056

Volume: 03 Issue: 10 | Oct -2016 www.irjet.net p-ISSN: 2395-0072

Queuing Theory in today’s world-An Overview
Vatsal Mishra1, Harshit Jain1, Ashish Nagra1, Naresh K2, Jagadeesh Kannan R3
1Student, BTech Computer Science and Engineering, VIT University, Vellore, TN, India
2Assistant Professor, School of Computer Science and Engineering, VIT University, Vellore, TN, India
3 Professor, School of Computer Science and Engineering, VIT University, Chennai, TN, India

---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - When all is said in done we don't prefer to hold 2.1 Example: Supermarket.
up. Be that as it may, decrease of the holding up time for the To what extent do clients need to hold up at the checkouts?
most part requires additional speculations. To choose whether What happens with the holding up 7 time amid pinnacle
or not to contribute, it is essential to know the impact of the hours? Are there enough checkouts?
speculation on the holding up time. So we require models and
strategies to investigate such circumstances. In this course we 2.2 Example: Production framework.
treat various basic queueing models. Consideration is paid to A machine produces distinctive sorts of items. What is the
techniques for the examination of these models, furthermore generation lead time of a request? What is the decrease
to uses of queueing models. Essential application regions of ahead of the pack time when we have an additional machine?
queueing models are creation frameworks, transportation and Would it be advisable for us to dole out needs to the
stocking frameworks, correspondence frameworks and data requests?
handling frameworks. Queueing models are especially helpful
for the outline of these framework regarding design, limits and 2.3 Example: Data correspondence.
control. In PC correspondence systems, for example, the Internet
Keywords: Queuing hypothesis, Queuing models, information bundles are transmitted over connections
Distributions, Probability Theory, Queuing Networks starting with one switch then onto the next. In every switch
approaching parcels can be cushioned when the approaching
1. INTRODUCTION interest surpasses the connection limit. Once the cushion is
A lining framework comprises of at least one servers that full, approaching bundles will be lost. What is the parcel
give administration or something to that affect to arriving delay at the switches? What is the division of parcels that
clients. Clients who touch base to discover all servers will be lost? What is a decent size of the cradle?
occupied for the most part go along with at least one (lines) 2.4 Example: Parking parcel.
before the servers, consequently the name lining
frameworks. There are a few regular cases that can be They will make another parking area before a grocery store.
depicted as lining frameworks, for example, bank employee How expansive would it be a good idea for it to be?
benefit, PC frameworks, producing frameworks, upkeep 2.5 Example: Assembly of printed circuit sheets.
frameworks, correspondences frameworks et cetera.
Mounting vertical parts on printed circuit sheets is done in a
Lining hypothesis is the scientific investigation of holding up gathering focus comprising of various parallel addition
lines, or lines. In lining hypothesis, a model is developed so machines. Every machine has a magazine to store parts.
that line lengths and holding up time can be anticipated. What is the creation lead time of the printed circuit loads up?
Lining hypothesis is for the most part considered a branch of By what means ought to the parts vital for the get together of
operations research in light of the fact that the outcomes are printed circuit sheets be separated among the machines?
regularly utilized when settling on business choices about
the assets expected to give an administration. 3. SINGLE QUEUEING NODES
Queueing hypothesis has its sources in research by Agner Single queueing hubs are generally portrayed utilizing
Krarup Erlang when he made models to depict the Kendall's documentation as a part of the frame A/S/C where
Copenhagen phone trade. The thoughts have since seen A depicts the time between landings to the line, S the span of
applications including media transmission, activity building, occupations and C the quantity of servers at the hub.
processing and the outline of industrial facilities, shops, Numerous hypotheses in lining hypothesis can be
workplaces and healing centers. demonstrated by lessening lines to numerical frameworks
known as Markov chains, initially depicted by Andrey
2. USE OF QUEUING Markov in his 1906 paper.
Cases Below we quickly portray a few circumstances in Agner Krarup Erlang, a Danish specialist who worked for the
which queueing is imperative. Copenhagen Telephone Exchange, distributed the main
paper on what might now be called lining hypothesis in

© 2016, IRJET | Impact Factor value: 4.45 | ISO 9001:2008 Certified Journal | Page 175
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 03 Issue: 10 | Oct -2016 www.irjet.net p-ISSN: 2395-0072

1909. He demonstrated the quantity of phone calls touching 3.4 Priority
base at a trade by a Poisson procedure and settled the
Clients with high need are served first. Need lines can be of
M/D/1 line in 1917 and M/D/k lining model in 1920. In
two sorts, non-pre-emptive (where an occupation in
Kendall's documentation:
administration can't be interfered) and pre-emptive (where
• M remains for Markov or memory less and implies an occupation in administration can be hindered by a higher-
landings happen as indicated by a Poisson procedure need work). No work is lost in either show.
• D remains for deterministic and means occupations 3.5 Shortest employment first
touching base at the line require a settled measure of
The following employment to be served is the one with the
administration
littlest size
• k depicts the quantity of servers at the queueing hub
3.6 Pre-emptive briefest occupation first
(k = 1, 2,...). On the off chance that there are a larger number
of occupations at the hub than there are servers then The following occupation to be served is the one with the
employments will line and sit tight for administration first littlest size
The M/M/1 line is a basic model where a solitary server 3.7 Shortest outstanding handling time
serves employments that land as indicated by a Poisson
The following employment to serve is the one with the
procedure and have exponentially dispersed administration
littlest outstanding handling necessity.
prerequisites. In a M/G/1 line the G remains for general and
shows a discretionary likelihood appropriation. The M/G/1 3.8 Service office
model was fathomed by Felix Pollaczek in 1930, an answer
• Single server: clients line up and there is stand out
later recast in probabilistic terms by Aleksandr Khinchin and
server
now known as the Pollaczek–Khinchine recipe.
• Parallel servers: clients line up and there are a few
After the 1940s queueing hypothesis turned into a region of
servers
research enthusiasm to mathematicians. In 1953 David
George Kendall comprehended the GI/M/k line and • Tandem line: there are numerous counters and
presented the current documentation for lines, now known clients can choose going where to line
as Kendall's documentation. In 1957 Pollaczek concentrated
on the GI/G/1 utilizing a basic condition. John Kingman gave 3.9 Customer's conduct of holding up
an equation for the mean holding up time in a G/G/1 line: • Balking: clients choosing not to join the line in the
Kingman's recipe. event that it is too long
The network geometric strategy and framework systematic • Jockeying: clients switch between lines in the event
strategies have permitted lines with stage sort appropriated that they think they will get served speedier by so doing
between entry and administration time dispersions to be
considered. • Reneging: clients leave the line on the off chance
that they have sat tight too ache for administration
Issues, for example, execution measurements for the M/G/k
line remain an open issue. 4. QUEUING NETWORKS

3. SERVER DISCIPLINES Systems of lines are frameworks in which various lines are
associated by client directing. At the point when a client is
Different planning strategies can be utilized at lining hubs: adjusted at one hub it can join another hub and line for
3.1 First in first out administration, or leave the system. For a system of m the
condition of the framework can be depicted by a m–
This standard expresses that clients are served each one in dimensional vector (x1,x2,...,xm) where xi speaks to the
turn and that the client that has been holding up the longest quantity of clients at every hub.
is served first.
The principal huge results around there were Jackson
3.2 Last in first out systems, for which a proficient item shape stationary
dispersion exists and the mean esteem analysis[24] which
This standard likewise serves clients each one in turn, yet
permits normal measurements, for example, throughput and
the client with the most brief holding up time will be served
visit times to be figured. In the event that the aggregate
first. Otherwise called a stack.
number of clients in the system stays consistent the system
3.3 Processor sharing is known as a shut system and has likewise been appeared to
have a product–form stationary dispersion in the Gordon–
Benefit limit is shared similarly between clients.
Newell hypothesis. This outcome was reached out to the
BCMP arrange, where a system with exceptionally broad
administration time, administrations and client directing is

© 2016, IRJET | Impact Factor value: 4.45 | ISO 9001:2008 Certified Journal | Page 176
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 03 Issue: 10 | Oct -2016 www.irjet.net p-ISSN: 2395-0072

appeared to likewise show an item frame stationary 4. Mayhew, Les; Smith, David (December 2006). Using
appropriation. The normalizing steady can be computed queuing theory to analyse completion times in
with the Buzen's calculation, proposed in 1973. accident and emergency departments in the light of
the Government 4-hour target. Cass Business
Systems of clients have additionally been researched, Kelly
School. ISBN 978-1-905752-06-5. Retrieved 2008-
systems where clients of various classes encounter diverse
05-20.
need levels at various administration hubs. Another kind of
system are G-organizes initially proposed by Erol Gelenbe in 5. Tijms, H.C, Algorithmic Analysis of Queues",
1993: these systems don't expect exponential time Chapter 9 in A First Course in Stochastic Models,
appropriations like the exemplary Jackson Network. Wiley, Chichester, 2003
5. Programming projects FOR SIMULATION/ANALYSIS
6. Kendall, D. G. (1953). "Stochastic Processes
The accompanying are the significant programming which Occurring in the Theory of Queues and their
are utilized for reenactment/examination Analysis by the Method of the Imbedded Markov
• Java Modeling Tools, a GPL suite of lining hypothesis Chain". The Annals of Mathematical
apparatuses written in Java Statistics. 24 (3):
338. doi:10.1214/aoms/1177728975. JSTOR 22362
• Queuing Package for GNU Octave 85.
• Discrete Event Simulation for Python
7. A.A. Markov, Extension of the law of large numbers
• Queuing Process Models in the Wolfram Language to dependent quantities, Izvestiia Fiz.-Matem.
Obsch. Kazan Univ., (2nd Ser.), 15(1906), pp. 135–
• PDQ programming bundle for R measurable
156 [Also [37], pp. 339–361].
processing
• SimEvents for MATLAB 8. "Agner Krarup Erlang (1878 - 1929) |
plus.maths.org". Pass.maths.org.uk. Retrieved 2013-
04-22.
6. CONCLUSIONS
9. Asmussen, S. R.; Boxma, O. J. (2009). "Editorial
With the information of likelihood hypothesis, information introduction". Queueing Systems. 63:
and yield models, and birth demise forms, it is conceivable to 1. doi:10.1007/s11134-009-9151-8.
determine various lining models, including however not
constrained to the ones we saw in this paper. Lining 10. Erlang, Agner Krarup (1909). "The theory of
hypothesis can be pertinent in numerous true circumstances. probabilities and telephone
For instance, seeing how to display a various server line conversations" (PDF). Nyt Tidsskrift for Matematik
could make it conceivable to decide what number of servers B. 20: 33–39. Archived from the original (PDF) on
are really required and at what wage so as to expand money 2011-10-01.
related proficiency. Then again maybe a lining model could
be utilized to concentrate on the life expectancy of the 11. Kingman, J. F. C. (2009). "The first Erlang century—
globules in road lights keeping in mind the end goal to better and the next". Queueing Systems. 63: 3–
see how habitually they should be supplanted. The 4. doi:10.1007/s11134-009-9147-4.
utilizations of lining hypothesis amplify well past holding up
in line at a bank. It might take some imaginative considering, 12. Pollaczek, F., Ueber eine Aufgabe der
yet in the event that there is any kind Wahrscheinlichkeitstheorie, Math. Z. 1930

REFERENCES 13. Whittle, P. (2002). "Applied Probability in Great
Britain". Operations Research. 50: 227–
177. doi:10.1287/opre.50.1.227.17792. JSTOR 3088
1. Sundarapandian, V. (2009). "7. Queueing 474.
Theory". Probability, Statistics and Queueing
Theory. PHI Learning. ISBN 8120338448. 14. Kendall, D.G.:Stochastic processes occurring in the
theory of queues and their analysis by the method
2. Lawrence W. Dowdy, Virgilio A.F. Almeida, Daniel of the imbedded Markov chain, Ann. Math. Stat.
A. Menasce. "Performance by Design: Computer
1953
Capacity Planning by Example".
15. Pollaczek, F., Problèmes Stochastiques posés par le
3. Schlechter, Kira (March 2, 2009). "Hershey Medical phénomène de formation d'une queue
Center to open redesigned emergency room". The
Patriot-News.

© 2016, IRJET | Impact Factor value: 4.45 | ISO 9001:2008 Certified Journal | Page 177
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 03 Issue: 10 | Oct -2016 www.irjet.net p-ISSN: 2395-0072

16. Kingman, J. F. C.; Atiyah (October 1961). "The single 27. Baskett, F.; Chandy, K. Mani; Muntz, R.R.; Palacios,
server queue in heavy traffic". Mathematical F.G. (1975). "Open, closed and mixed networks of
Proceedings of the Cambridge Philosophical queues with different classes of customers". Journal
Society. 57 (4): of the ACM. 22 (2): 248–
902. doi:10.1017/S0305004100036094. JSTOR 298 260. doi:10.1145/321879.321887.
4229.
28. Buzen, J. P. (1973). "Computational algorithms for
17. Ramaswami, V. (1988). "A stable recursion for the closed queueing networks with exponential
steady state vector in markov chains of m/g/1 servers" (PDF). Communications of the ACM. 16 (9):
type". Communications in Statistics. Stochastic 527. doi:10.1145/362342.362345.
Models. 4: 183–
188. doi:10.1080/15326348808807077. 29. Kelly, F. P. (1975). "Networks of Queues with
Customers of Different Types". Journal of Applied
18. Penttinen A., Chapter 8 – Queueing Systems, Lecture Probability. 12 (3): 542–
Notes: S-38.145 - Introduction to Teletraffic Theory. 554. doi:10.2307/3212869. JSTOR 3212869.

19. Harchol-Balter, M. (2012). "Scheduling: Non- 30. Gelenbe, Erol (Sep 1993). "G-Networks with
Preemptive, Size-Based Policies". Performance Triggered Customer Movement". Journal of Applied
Modeling and Design of Computer Systems. Probability. 30 (3): 742–
p. 499. doi:10.1017/CBO9781139226424.039. ISBN 748. doi:10.2307/3214781. JSTOR 3214781.
9781139226424.
31. Bobbio, A.; Gribaudo, M.; Telek, M. S. (2008).
20. Harchol-Balter, M. (2012). "Scheduling: "Analysis of Large Scale Interacting Systems by
Preemptive, Size-Based Policies". Performance Mean Field Method". 2008 Fifth International
Modeling and Design of Computer Systems. Conference on Quantitative Evaluation of Systems.
p. 508. doi:10.1017/CBO9781139226424.040. ISBN p. 215. doi:10.1109/QEST.2008.47. ISBN 978-0-
9781139226424. 7695-3360-5.

21. Harchol-Balter, M. (2012). "Scheduling: SRPT and 32. Bramson, M. (1999). "A stable queueing network
Fairness". Performance Modeling and Design of with unstable fluid model". The Annals of Applied
Computer Systems. Probability. 9 (3):
p. 518. doi:10.1017/CBO9781139226424.041. ISBN 818. doi:10.1214/aoap/1029962815. JSTOR 26672
9781139226424. 84.

22. Jackson, J. R. (1957). "Networks of Waiting 33. Chen, H.; Whitt, W. (1993). "Diffusion
Lines". Operations Research. 5 (4): 518– approximations for open queueing networks with
521. doi:10.1287/opre.5.4.518. JSTOR 167249. service interruptions". Queueing Systems. 13 (4):
335. doi:10.1007/BF01149260.
23. Jackson, James R. (Oct 1963). "Jobshop-like
Queueing Systems". Management Science. 10 (1): 34. Yamada, K. (1995). "Diffusion Approximation for
131– Open State-Dependent Queueing Networks in the
142. doi:10.1287/mnsc.1040.0268. JSTOR 2627213. Heavy Traffic Situation". The Annals of Applied
Probability. 5 (4):
24. Reiser, M.; Lavenberg, S. S. (1980). "Mean-Value 958. doi:10.1214/aoap/1177004602. JSTOR 22451
Analysis of Closed Multichain Queuing 01.
Networks". Journal of the ACM. 27 (2):
313. doi:10.1145/322186.322195. 35. Bertoli, Marco; Casale, Giuliano; Serazzi, Giuseppe
(2009). "JMT: performance engineering tools for
25. Van Dijk, N. M. (1993). "On the arrival theorem for system modeling". SIGMETRICS Perform. Eval.
communication networks". Computer Networks and Rev. 36: 10–15. doi:10.1145/1530873.1530877.
ISDN Systems. 25 (10): 1135–
2013. doi:10.1016/0169-7552(93)90073-D. 36. Marzolla, Moreno. "The queueing Package" (PDF).
Retrieved 31 July 2014.
26. Gordon, W. J.; Newell, G. F. (1967). "Closed Queuing
Systems with Exponential Servers". Operations 37. Müller, Klaus; Vignaux, Tony. "SimPy
Research. 15 (2): Acknowledgements". Retrieved 24 May 2016.
254. doi:10.1287/opre.15.2.254. JSTOR 168557.

© 2016, IRJET | Impact Factor value: 4.45 | ISO 9001:2008 Certified Journal | Page 178
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 03 Issue: 10 | Oct -2016 www.irjet.net p-ISSN: 2395-0072

38. "Queueing Processes—Wolfram Language
Documentation". reference.wolfram.com.
Retrieved 2016-06-23.

39. "PDQ-R: {\em Pretty Damn Quick} for R Statistical
Computing". www.perfdynamics.com.
Retrieved 2016-06-23.

40. "Queuing - MATLAB &
Simulink". www.mathworks.com. Retrieved 2016-
06-23.

© 2016, IRJET | Impact Factor value: 4.45 | ISO 9001:2008 Certified Journal | Page 179